 From theCUBE Studios in Palo Alto in Boston, bringing you data-driven insights from theCUBE and ETR. This is Breaking Analysis with Dave Vellante. In June of this year, Snowflake filed a confidential document suggesting that it would do an IPO. Now of course, everybody knows about it, found out about it, and it had a $20 billion valuation. So many in the community, in the investment community and so forth are excited about this IPO. It could be the hottest one of the year. And we're getting a number of questions from investors and practitioners and the entire Wikibon, ETR, and CUBE community. So welcome everybody, this is Dave Vellante. This is CUBE Insights powered by ETR. And in this Breaking Analysis, we're going to unpack five critical questions around Snowflake's IPO or pending IPO. And with me to discuss that is Eric Bradley. He's the chief engagement strategist at ETR. He's also the managing director of Venn. Eric, thanks for coming on, great to see you as always. Great to see you too, always enjoy being on the show, thank you. Now for those of you who don't know Eric, Venn is a roundtable that he hosts and he brings in CIOs, IT practitioners, CISOs, data experts, and they have an open and frank conversation, but it's private to ETR clients, but they know who the individual is, what their role is, what their title is, et cetera. And it's a kind of an ask me anything and I participated in one of them this past week outstanding and we're going to share with you some of that, but let's bring up the agenda slide if we can here. And these are really some of the questions that we're getting from investors and others in the community. There's really five areas that we want to address. The first is what's happening in this enterprise data warehouse marketplace. And the second thing, it's kind of a 1A, is what about the legacy EDW players like Oracle and Teradata and Nutiza? The third question we get a lot is, can Snowflake compete with the big cloud players, Amazon, Google, Microsoft, they're right there in the heart and the thick of things there. And then what about that multi-cloud strategy? Is that viable? How much of a differentiator is that? And then we get a lot of questions on the TAM and how, meaning the total available market, how big is that market? Does it justify the valuation for Snowflake? Now, Eric, you've been doing this now, you've run a couple of vans, you've been following this, you've done some other work that you've done with Eagle Alpha. What's your initial takeaway from all this work that you've been doing? Yeah, sure, so my first take on Snowflake was about two and a half years ago. I actually hosted them for one of my van interviews and my initial thought was impressed, so impressed. They were talking at the time about their ability to kind of make ease of use of a multi-cloud strategy. At the time, although I was impressed, I did not expect the growth and the hypergrowth that we have seen now. But looking at the company and its current iteration, I understand where the hype is coming from. I mean, a 12 and a half billion private valuation in the last round, the least confidential IPO anyone's ever seen with a 15 to $20 billion valuation coming out, which is more than Teradata, Mongo, and Cloudera combined, it's a great question. Obviously, the success to this point is warranted, but we need to see what they're going to be able to do next. I think the agenda you laid out is a great one and I'm looking forward to getting into some of those details. So let's start with what's happening in the marketplace and let's pull up a slide that I very much love to use. It's the classic X, Y on the vertical axis here. We show net score and I remember folks, net score is an indicator of spending momentum, ETR every quarter does a like clockwork survey where they're asking people, essentially, are you spending more or less? They subtract the less from the more and it comes up with a net score. It's more complicated than that, but like NPS, it's a very simple and reliable methodology. That's the vertical axis and the horizontal axis is what's called market share. Market share is the pervasiveness within the data set. So it's calculated by the number of mentions of the vendor divided by the number of mentions within that sector. And what we're showing here is the EDW sector. And we've pulled out a few companies that I want to talk about. So the big three obviously are Microsoft, AWS and Google. And you can see Microsoft has a huge presence far to the right, AWS very, very strong, a lot of redshift in there. And then you're pretty high on the vertical axis and then Google not as much share, but very solid in that close to 60% net score. And then you can see above all of them from a vertical standpoint is Snowflake with a 77.5% net score. You can see them in the upper right there in the green. One of the highest, Eric, in the entire data set. So let's start with some sort of initial comments on the big guys and Snowflakes, your thoughts. Sure, just first of all to comment on the data. What we're showing there is just the data warehousing sector, but Snowflake's actual net score is that high amongst the entire universe that we follow. Their data strength is unprecedented and it really, we have forward looking spending intentions so this bodes very well for them. Now, what you did say very accurately is there's a difference between their spending intentions on a net revenue level compared to AWS, Microsoft. No one's saying that this is an apples to apples comparison when it comes to actual revenue. So we have to be very cognizant of that. There's domination quite frankly from AWS and from Azure and Snowflake is a necessary component for them not only to help facilitate a multi-cloud, but look what's happening right now in the US Congress. We have these tech leaders being grilled on their actual dominance and one of the main concerns they have is the amount of data that they're collecting. So I think the environment is right to have another player like this. I think Snowflake really has a lot of longevity and our data is supporting that and the commentary that we hear from our end users, the people that take the survey are supporting that as well. Okay, and let's stay on this XY slide for a moment and I want to just pull out a couple of other comments here because one of the questions we're asking is wither the legacy EDW players. So we've got in here IBM Oracle, you can see Teradata and then Hortonworks and MapR and we're going to talk a little bit where Hortonworks of course is now cloud era. We're going to talk a little bit about Hadoop and some of the data lakes. So you can see there, they don't have nearly the net score momentum. Oracle obviously has a huge install base and is investing quite frankly in R&D and doing exadata and has its own cloud. So it's got a lock on, it's customers and if it keeps investing and adding value, it's not going away. IBM with Natiza, there's really been some questions around their commitment to that base and I know that a lot of the folks in the Venns that we've talked to Eric have said, well, we're replacing Natiza or Frank Slutman has been very vocal about going after Teradata. And then we're going to talk a little bit about the Hadoop space, but can you summarize for us your thoughts and your research and the commentary from your community? What's going on with the legacy guys? Are these guys cooked? Can they hang on? What's your take? Sure, we focus on this quite a bit actually. So I'm going to talk about it from the data perspective first and then we'll go into some of the commentary in the panel. You even joined one yesterday, you know that it was touched upon. But first on the data side, what we're noticing and capturing is a widening bifurcation between these cloud native and the legacy on-prem. It is undeniable. There is nothing that you can really refute. The data is concrete and it is getting worse. That gap is getting wider and wider and wider. Now, the one thing I will say is, you know, nobody's going to rip out their legacy applications tomorrow. It takes years and years. So when you look at ATERRA data, right? Their market cap is only 2 billion, 2.3 billion. How much revenue growth do they need to stay where they are? Not much, right? No one's expecting them to grow 20%, which is what you're seeing on the left side of that screen. So when you look at the legacy versus the cloud native, there is very clear direction of what's happening. The one thing I would note from the data perspective is if you switched from net score or adoptions and you went to flat spending, you suddenly see Oracle and ATERRA data move over to that left a little bit because, again, what I'm trying to say is, I don't think they're going to catch up. No, but I also don't think they're going away tomorrow. You know, these have large install bases. They have relationships. Now, to kind of get into what you were saying about each particular one, IBM, they shut down the Tesla. They shut it down and then they brought it back to life. How does that make you feel if you're the head of data architecture or you're dev ops and you're trying to build an application for our large company? I'm not going back to that. There's absolutely no way. ATERRA data, on the other hand, is known to be incredibly stable. They are known to just not fail. If you need to kind of re-architect or you do a migration, they work. ATERRA data also has a lot of compliance built in. So if you're a financials, if you have a regulated business or industry, there's still some data sets that you're not going to move up to the cloud, whether it's a PII compliance or financial reasons. Some of that stuff is still going to live on-prem. So ATERRA data still has a very good niche. And from what we're hearing from our panels, this is a direct quote, if you don't mind me looking off screen for one second, but this is a great one, basically said ATERRA data is the only one for the legacy camp who was putting up a fight and not giving up. Basically from a CIO perspective, the rest of them aren't an option anymore. But ATERRA data is still fighting. And that's great to hear. They have their own data as a service offering. And listen, they're a small, small market cap compared to these other companies we're talking about. But to summarize, the data is very clear. There is a widening bifurcation between the two camps. I do not think legacy will catch up. I think all net new workloads are moving to data as a service, moving to cloud data, moving to hosted, but there are still going to be some existing legacy on-prem applications that will be supported with these older databases. And of those, Oracle and ATERRA data are still viable options. I totally agree with you. And my colleague, David Floyer, is actually quite high on ATERRA data vantage because he really does believe that a key component, we're going to talk about the TAM in a minute, but a key component of the TAM, he believes must include the on-premises workloads. Frank Slutman's been very clear. We're not doing on-prem. We're not doing this halfway house. And so that's an opportunity for companies like TeraData, certainly Oracle, I would put in that camp is putting up a fight. Vertic is another one. They're very small, but another one that's sort of battling it out from the old MPP world. But that's great. Let's go into some of the specifics. Let's bring up here some of the specific commentary and that we've curated here from the roundtables. I'm going to go through these and then ask you to comment. The first one is just, I mean, people are obviously very excited about Snowflake. It's easy to use, you know, the whole thing zero to Snowflake in 90 minutes, but Snowflake is synonymous with cloud native data warehousing. There are no equals. We heard that a lot from your Venn panelists. We certainly did. There was even more euphoria around Snowflake than I expected when we started hosting these series of data warehousing panels. And this particular gentleman that said that happens to be the global head of data architecture for a Fortune 100 financials company. And you mentioned earlier that we did a report alongside Eagle Alpha. And we noticed that among Fortune 100 companies that are also using the big three public cloud companies, Snowflake is growing market share faster than anyone else. They are positioned in a way where even if you're aligned with Azure, even if you're aligned with AWS, if you're a large company, they are gaining share right now. So that particular gentleman's comment was very interesting. He also made a comment that said, Snowflake is the person who championed the idea that data warehousing is not dead yet. Use that old Monty Python line, you know, not dead yet. And, you know, back in the day where the Hadoop came along and the data lakes turned into a data swamp and everyone said, we don't need warehousing anymore. Well, that turned out to be a head fake, right? Hadoop, you know, was an interesting technology, but it's a complex technology and it ended up not really working the way people wanted. I think Snowflake came in at that point at an opportune time and said, no, data warehousing isn't dead. We just have to separate the compute from the storage layer and look at what I can do. That increases flexibility, security. It gives you that ability to run across multi-cloud. So honestly, the commentary has been nothing but positive. We can get into some of the commentary about people thinking that there's competition catching up to what they do, but there is no doubt that right now, Snowflake is the name when it comes to data as a service. The other thing we heard a lot was ETL is going to get completely disrupted. You sort of embedded ETL. You heard one panelist say, well, it's interesting to see that guys like Informatica are talking about how fast they can run inside of Snowflake, but Snowflake is making that easy that data prep is sort of part of the package. And so that does not bode well for ETL vendors. Does not, right? So ETL is a legacy of on-prem databases, and even when Hadoop came along, it still needed that extra layer to kind of work with the data, but this is really, really disrupting them. Now, the Snowflake's credit, they partner well. All the ETL players are partnered with Snowflake. They're trying to play nice with them, but the writings on the wall, as more and more of this application and I'm sorry, workloads move to the cloud, you don't need the ETL layer. Now, obviously that's gonna affect talent and Informatica the most. We had a recent comment that said, this was a CIO who basically said, the most telling thing about the ETL players right now is every time you speak to them, all they talk about is how they work in a Snowflake architecture. That's their only metric that they talk about right now. And he said, that's very telling that he basically used it as their, it's their existential identity to be part of Snowflake. If they're not, they don't exist anymore. So it was interesting to have sort of a philosophical comment brought up in one of my roundtables, but that's how important playing nice and finding a niche within this new data as a service is for ETL. But to be quite honest, they might be going the same way of, okay, let's figure out our niche on the on-prem workloads that are still there. I think over time, we might see them maybe as an M&A possibility, whether it's Snowflake or one of these new up-and-comers, kind of bring them in and sort of take some of the technology that's useful and layer it in. But as a large market cap, solo existing niche, I just don't know how long ETL is for this world. Now, yeah, I mean, you're right, they've been for the marketing, they're not fighting fashion, but really there are some challenges there. Now, there were some contrarians in the panel and they signal some potential icebergs ahead. And I guarantee you're going to see this in Snowflake's red herring when we actually get it. We're going to see all the risks. One of the comments, I'll mention the two and then we can talk about it. Their engineering advantage will fade over time. Essentially, we're saying that people are going to copycat and we've seen that. And the other point is, hey, we might see some similar things that happen to Hadoop, the public cloud players giving away these offerings at zero cost, essentially marginal cost of adding in other services is near zero. So the cloud players will use their heft to compete your thoughts. Yeah, you know, first of all, that's one of the reasons I love doing panels, right? Because we had three gentlemen on this panel that all had nothing but wonderful things to say, but you always get one. And this particular person is a CTO of a well-known online public travel agency. We'll put it that way. And he said, you know, I'm going to be the contrarian here. I have seven different technologies from private companies that do the same thing that I'm evaluating. So that's the pressure from Bob behind, right? The technology, they're going to catch up. Right now, Snowflake has the best engineering, which interestingly enough, they took a lot of that engineering from IBM and Teradata. If you actually go back and look at it, which was brought up in our panel as well. He said, however, the engineering will catch up. They always do. Now from the other side, they're getting squeezed because the big cloud players just say, hey, we can do this too. I can bundle it with all the other services I'm giving you. And I could squeeze it, it'd be pretty much give it away at a cost. So I do think that there is a very valid concern when you come out with a $20 billion IPO evaluation, you need to warrant that. And when you see competitive pressures from both sides from private emerging technologies and from the more dominant public cloud players, you're going to get squeezed there a little bit. And if pricing gets squeezed, you know, it's going to be very, very important for Snowflake to continue to innovate. That comment you brought up about possibly being the next Cloudera was certainly the best soundbite that I got. And I'm going to use it as clickbait in future articles because I think everyone who starts looking to buy Snowflake stock and they see that, they're going to need to need to take a look. But I would take that with a grain of salt. I don't think that's happening anytime soon. But what that particular CTO was referring to was, if you don't innovate, the technology itself will become commoditized and he believes that this technology will become commoditized. So therefore, Snowflake has to continue to innovate. They have to find other layers to bring in, whether that's through their massive war chest of cash they're about to have and M&A, whether that's them buying analytics company, whether that's them buying an ETL layer, finding a way to provide more value as they move forward is going to be very important for them to justify this valuation going forward. Yeah, and I want to comment on that, the Cloudera, Hortonworks, MapR piece, Hadoop, et cetera. I mean, there are dramatic differences, obviously. I mean, that whole space was so hard, very difficult to stand up. You needed, you know, science project guys and lab coach to do it. It was very services intensive. As well, companies like Cloudera had to fund all these open source projects and it really squeezed their R&D. I think Snowflake is much more focused and you mentioned some of the background of their engineers, of course, Oracle guys as well. However, you will see Amazon's going to trot out a ton of customers using their RA3 managed storage and their flash, they think this is the DC2 piece. They have a ton of action in the marketplace because it's just so easy. It's interesting, one of the comments you asked this yesterday with regard to separating compute from storage, which of course is Snowflake's, they basically invented it, it was one of their claims to fame. The comment was what AWS has done to separate compute from storage for Redshift is largely a bolt on, which I thought that was an interesting comment. I've had some other comments, my friend George Gilbert said, hey, despite claims to the contrary, AWS still hasn't separated storage from compute, what they have is really primitive. We got to dig into that some more, but you're seeing some data points that suggest there's copycatting going on, may not be as functional, but at the same time, Eric, like we were saying, good enough is maybe good enough in this space. Yeah, and especially with the enterprise, right? You see what Microsoft has done. Their technology is not as good as all the niche players, but it's good enough, and I already have a Microsoft license, so why am I gonna move off of it? But I wanna get back to that comment you mentioned too about that particular gentleman who made that comment about Redshift, their separation is really more of a bolt-on than a true offering. It's interesting because I know who these people are behind the scenes, and he has a very strong relationship with AWS. So it was interesting to me that in the panel yesterday, he said he switched from Redshift to Snowflake because of that and some other functionality issues. So there is no doubt from the end users that are buying this, and he's again a Fortune 100 financial organization, not the same one we mentioned earlier, a different one, but again, a Fortune 100 well-known financial organization. He switched from AWS to Snowflake. So there is no doubt that right now they have the technological lead, and when you look at our ETR data platform, you know we have that adoption reasoning slide that you show. When you look at it, the number one reason that people are adopting Snowflake is their feature set and technological lead. They have that lead now. They have to maintain it. Now, another thing to bring up on this to think about is when you have large data sets like this and as we're moving forward, you need to have machine learning capabilities layered into it, right? So they need to make sure that they're playing nicely with that. And now you could go open source with the Apache Suite, but Google is doing so well with BigQuery and so well with their machine learning aspects. And although they don't speak enterprise well, they don't sell to the enterprise well, that's changing. I think they're somebody to really keep an eye on because their machine learning capabilities that are layered into BigQuery are impressive. Now, of course, Microsoft Azure has Databricks. They're layering that in, but this is an area where I think you're going to see maybe what's next. You have to have machine learning capabilities out of the box if you're gonna do data as a service. Right now, Snowflake doesn't really have that. Some of the other ones do. So I had one of my guest panelists basically say to me because of that, they ended up going with Google BigQuery because he was able to run a machine learning algorithm within hours of getting set up, within hours. And he said that that kind of capability out of the box is what people are going to have to use going forward. So that's another thing we should dive into a little bit more. Well, let's get into that right now. Let's bring up the next slide, which shows NetScore. Remember, this is spending momentum across the major cloud players and plus Snowflake. So you've got Snowflake on the left, Google, AWS and Microsoft. And it's showing three survey time frames. Last October, April 20, which is right in the middle of the pandemic. And then the most recent survey, which has just taken place this month in July. And you can see Snowflake, very, very high scores, actually improving from the last October survey. You know, Google lower NetScore's, but still very strong. I'm going to come back to that and pick up on your comments. AWS dipping a little bit. And I think what's happening here, we saw this yesterday with AWS as results, 30% growth, awesome, slightly miss, on the revenue side for AWS. But look, I mean, massive, and they're so exposed to so many industries. So some of their industries have been pretty hard hit. Microsoft, pretty interesting. A little softness there, but one of the things I wanted to pick up on, Eric, when you're talking about Google and BigQuery, and it's, you know, ML out of the box was what we heard from a lot of the Venn participants. There's no question about it that Google technically, I would say is one of Snowflake's biggest competitors because it's cloud native. Remember, AWS did a one-time license deal with Power Excel and had to sort of refactor the thing to be cloud native. And of course we know what's happening with Microsoft. They basically were on-prem and then they put stuff in the cloud. Then all the updates happen in the cloud and then they push to on-prem. But there's, you know, they have that, what Frank Slutman calls that halfway house. But BigQuery, no question technically is very, very solid. But again, you see Snowflake right now anyway, outpacing these guys in terms of momentum. Snowflake is outpacing everyone across our entire survey universe. It really is impressive to see. And one of the things that they have going for them is they can connect all three. It's that multi-cloud ability, right? That portability that they bring to you is such an important piece for today's modern CIOs or data architects. They don't want vendor lock-in. They are afraid of vendor lock-in. And this ability to make their data portable and to do that with ease and the flexibility that they offer is a huge advantage right now. However, I think you're 100% right. Google has been so focused on the engineering side and never really focusing on the enterprise sales side. That is why they're playing catch-up. I think they can catch up. They are bringing in some really important enterprise sales people with experience. They're starting to learn how to talk to enterprise, how to sell, how to support. And nobody can really doubt their engineering. How many open sources have they given us, right? They invented Kubernetes in the entire container space. No one's really gonna compete with them on that side if they learn how to sell it and support it. Yeah, right now they're behind. They're a distant third. Don't get me wrong. From a pure hosted ability, AWS is number one. Microsoft Azure sometimes looks like it's number one, you have to recognize that a lot of that is because of simply their hosted 365. It's a SaaS app. It's not a true cloud type of infrastructure as a service. But Google is a distant third, but their technology is really, really great and their ability to catch up is there. And like you said, in the panels, we were hearing a lot about their machine learning capabilities right out of the box. And that's where this is going. What's the point of having this huge data if you're not gonna be supporting the new application architecture and all of those applications require machine learning? Awesome, so we're, and I totally agree with what you're saying about Google. They just don't have it figured out how to sell the enterprise yet. And 100%, AWS has the best cloud. I mean, hands down. But the very, very competitive market as we heard yesterday in front of Congress. Now, we're on the point about, can Snowflake compete with the big cloud players? I want to show one more data point. So let's bring up, this is the same chart as we showed before, but it's new adoptions. And this is really telling. You can see Snowflake, with 34% in the yellow, new adoptions, down, yes, from previous surveys, but still significantly higher than the other players. Interesting to see Google showing momentum on new adoptions, AWS down on new adoptions. And again, exposed to a lot of industries that have been hard hit and Microsoft actually quite low on new adoptions. So this is very impressive for Snowflake. And I want to talk about the multi-cloud strategy now, Eric. This came up a lot. The Venn participants who were sort of fans of Snowflake said three things. It was really the flexibility, the security, which was really interesting to me. And a lot of that had to do with the flexibility, the ability to easily set up roles and not have to waste a lot of time wrangling. And then the third was multi-cloud. And that was really something that came through heavily in the Venn, didn't it? It really did. And again, I think it just comes down to, I don't think you can ever overstate how afraid these guys are of vendor lock-in. They can't have it, they don't want it. And it's also just, it's best practice to make sure your sensitive information is being kind of spread out a little bit. We all know that people don't trust Bezos. So if you're in certain industries, you're not gonna use AWS at all, right? So yeah, this ability to have your data portability through multi-cloud is the number one reason I think people start looking at Snowflake. And to go to your point about the adoptions, it's very telling and it bodes well for them going forward. Most of the things that we're seeing right now are net new workloads. So let's go again back to the legacy side that we were talking about the Teradata's, IBM's, Oracle's. They still have the monolithic applications and the data that needs to support that, right? Like an old ERP type of thing. But anyone who's now building a new application, bringing something new to market, it's all net new workloads. There is no net new workload that is gonna go to SAP or IBM. It's not gonna happen. The net new workloads are going to the cloud. And that's why when you switch from net score to adoption, you see Snowflake really stand out because this is about new adoption for net new workloads and that's really where they're driving everything. So I would just say that as this continues, as data as a service continues, I think Snowflake's only gonna gain more and more share. For all the reasons you stated. Now, get back to your comment about security. I was shocked by that. I really was. I did not expect these guys to say, oh no, Snowflake Enterprise Security? Not a concern. So two panels ago, a gentleman from Fortune 100 Financials said, listen, it's very difficult to get us to sign off on something for security. Snowflake has passed it. It is enterprise ready and we are going full steam ahead. Once they got that go ahead, there was no turning back. We gave it to our DevOps guys. We gave it to everyone and said, run with it. So, you know, when a company that's big, I believe their fortune rank is 28. So when a company that big says, yeah, you've got the green light that we were okay with the internal compliance aspect. We're okay with the security aspect. This gives us multi-cloud portability. This gives us flexibility ease of use. Honestly, there's a really long runway ahead for Snowflake. Yeah, so the big question I have around the multi-cloud piece and I totally, and I've said, I've been on record saying, look, if you're going to looking for an agnostic multi-cloud, you're probably not going to go with the cloud vendor. You know, so, and so, but I've also said that I think multi-cloud to date anyway has largely been a symptom as opposed to a strategy, but that's changing to your point about lock in. And also I think people are maybe looking at doing things across clouds. But I think that certainly it expands Snowflake's TAM, and we're going to talk about that because they support multiple clouds and they're going to be the best at that. That's a mandate for them. The question I have is how much complex joining are you going to be doing across clouds? And is that something that is just going to be too latency intensive? Is that really Snowflake's expertise? They're really trying to build that data layer. You're probably going to maybe use some kind of Postgres database for that. I don't know, I need to dig into that, but that would be an opportunity from a TAM standpoint. I just don't know how real that is. Yeah, unfortunately, I'm going to just be honest with this one. I don't think I have great expertise there and I wouldn't want to lead anyone the wrong direction, but from what I've heard from some of my, you know, then interview subjects, this is happening. So the data portability needs to be agnostic to the cloud. I do think that what you're saying are they're going to be real complex kind of workloads and applications? Yes, the answer is yes. And I think a lot of that has to do with some of the container architecture as well, right? If I can just pull data from one spot, spin it up for as long as I need it and then just get rid of that container, that ethereal layer of compute, it doesn't matter where the cloud lies. It really doesn't. I do think that multi-cloud is the way of the future. I know that the container workloads right now in the enterprise are still very small. I've heard people say like, yeah, I've kicked into tires, we got 5%. That's going to grow. And if Snowflake can make themselves an integral part of that, then yes, I think that's one of those things where remember the guy said, Snowflake has to continue to innovate. They have to find a way to grow this TAM. This is an area where they can do so. I think you're right about that. But as far as my expertise on this one, I'm going to be honest with you and say, I don't want to answer incorrectly. So you and I need to dig in a little bit on this one. Yeah, as it relates to question four, what's the viability of Snowflake's multi-cloud strategy? I'll say unquestionably supporting multiple clouds, very viable. Whether or not portability across clouds, multi-cloud joins, et cetera, TBD. So we'll keep digging into that. The last thing I want to focus on here is the last question to Snowflake's TAM justify its $20 billion valuation. And you think about the data pipeline. You go from data acquisition to data prep. I mean, that really is where Snowflake shines. And then of course there's analysis, you got to bring in MI or AI and ML tools. That's not Snowflake's strength. And then you're obviously preparing that, serving that up to the business, visualization. So there's potential adjacencies that they could get into that they may or may not decide to. But so we put together this next chart, which is kind of the TAM expansion opportunity. And I just want to briefly go through it. We publish this stuff so you can go and look at all the fine print, but it kind of starts with the data lake disruption. You call the data swamp before, the Hadoop, no schema on right, basically the ROI of Hadoop became reduction of investment, as my friend Avi Mehta would say. But so they're kind of disrupting that data lake which really was a failure. And then really going after that enterprise data warehouse, which is kind of a, I have it here as a 10 billion. It's actually bigger than that. It's probably more like a $20 billion market. I'll update this slide. And then really what Snowflake is trying to do is be data as a service. A data layer across data stores, across clouds, really make it easy to ingest and prepare data and then serve the business with insights. And then ultimately this huge TAM around automated decision-making, real-time analytics, automated business processes. I mean, that is potentially an enormous market. We got a couple a hundred billion. I mean, just huge. Your thoughts on their TAM? I agree. I'm not worried about their TAM. And one of the reasons why is I mentioned before, they are coming out with a whole lot of cash. This is going to be a red hot IPO. They are going to have a lot of money to spend and look at their management team. Who is leading the way? A very successful, wise, intelligent, acquisitive type of CEO. I think there is going to be M&A activity and I believe that M&A activity is going to be 100% for the mindset of growing that TAM. The entire world is moving to data as a service. So let's take as a backdrop. I'm gonna go back to the panel we did yesterday. The first question we asked was there was an understanding or a theory that when the virus pandemic hit, people wouldn't be taking on any sort of net new architecture. They're like, okay, I have tarry data. I have IBM. Let's just make sure the lights are on. Let's stick with it. Every single person I've asked, there's just now eight different experts said to us, oh no, oh no, no. There's the virus pandemic, the shift from work from home, everything we're seeing right now has only accelerated and advanced our data as a service strategy in the cloud. We are building for scale, adopting cloud for data initiatives. So across the board, they have a great backdrop. So that's going to only continue, right? This is very new. We're in the early innings of this. So for their TAM, that's great because that's the core of what they do. Now on top of it, you mentioned a type of things about, yeah, right now they don't have great machine learning. That can easily be acquired and built in. Right now they don't have an analytics layer. I for one would love to see these guys talk to Alteryx. Alteryx is red hot. We're seeing great data and great feedback on them. If they could do that business intelligence and that analytics layer on top of it, the entire suite as a service, I mean, come on. Their TAM is expanding, in my opinion. Yeah, your point about their leadership is right on. I interviewed Frank Slutman, written the heart of the pandemic. And he said, I'm investing in engineering, almost sight unseen, more of a circumspect around sales. But I will caution people that a lot of people I think see what Slutman did with service now. And he came into service now, I have to tell you, they didn't have their unit economics right. They didn't have their sales model and marketing model. He cleaned that up, took it from 100, 120 million to 1.2 billion and really did an amazing job. People are looking for a repeat here. This is a totally different situation. Service now drove a truck through BMC's install base and with IT help desk and then created this brilliant TAM expansion let's land and expand model. This is much different here. And Slutman also told me that he's a situational CEO. He doesn't have a playbook. And so that's what is most impressive and interesting about this. He's now up against the biggest competitors in the world, AWS, Google and Microsoft and dozens of other smaller startups that have raised a lot of money. Look at the company like Yellowbrick. They've raised 180 million dollars. They got a great team, Google, IBM, et cetera. So it's going to be really, really fun to watch. I'm super excited, Eric, but I'll tell you the data right now suggest they've got a great tailwind and if they can continue to execute this is going to be really fun to watch. Yeah, certainly. I mean, when you come out and you are as impressive as Snowflake is you get a target on your back. There's no doubt about it, right? So we said that, you know they basically created the data as a service. That's going to invite competition. There's no doubt about it. And Yellowbrick is one that came up in the panel yesterday about one of our CIOs were doing a proof of concept with them. We had about seven others mentioned as well that are startups that are in this space. However, none of them, despite their great valuation and their great funding are going to have the kind of money and the market lead that Slutman's going to have with Snowflake as this comes out. And what we're seeing in Congress right now with some antitrust scrutiny around the large data that's being collected by AWS, Azure, Google. You know, I'm not going to bet against this guy either. Right now, I think he's got a lot of opportunity. There's a lot of additional layers and because he can, you know basically develop this as a sweet service. I think there's a lot of a great opportunity ahead for this company. Yeah, and I guarantee that he's got his, he understands well as a customer acquisition cost and the lifetime value of the customer, the retention rates. Those are all things that he and Mike Scarpelli as CFO learned at service now, not learned, perfected. And so, well, Eric, really great conversation. Awesome data. It's always a pleasure having you on. Thank you so much, my friend. I really appreciate it. I appreciate talking to you too. We'll do it again soon and stay safe everyone out there. All right, and thank you for watching everybody. This episode of Cube Insights, powered by ETR. This is Dave Vellante and we'll see you next time.